Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Even with annotations present (readOnlyHint:false, idempotentHint:true), the description adds substantial behavioral context: the 50-watch hard limit per Bearer key, the authentication requirement, signed callback semantics, and duplicate/unsafe rejection behavior. It richly reinforces the idempotentHint by explaining exactly how duplicate registration is handled, and clarifies that despite being a write operation, it intentionally avoids immediate spec ingestion—semantics no structured field could capture.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Three dense sentences in under 65 words cover purpose, authentication, limits, success criteria, and exclusions. The structure is front-loaded with the primary verb-object, then constraints, then a final clarifying negative. Every clause earns its place with no filler, redundancy, or schema repetition.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's moderate complexity (a stateful subscribe action) plus the presence of an output schema, the description fully covers the open questions: side effects are explicit (watch creation), failure modes are disclosed (duplicate/unsafe rejection), limits are quantified (50 per key), and the temporal scope ('after later spec changes') clarifies it won't fire retroactively.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline of 3 is appropriate; the schema already provides format (uri), constraints, and clear examples for both parameters. The description marginally enriches webhookUrl by framing the poll-vs-callback tradeoff, but it largely restates what the schema fields already declare. No parameters are undocumented, so nothing is missing, but the description adds little parameter-specific insight beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The opening phrase 'Create a persistent specification watch for one service' uses a specific verb and resource with clear scope, and the 'This does not ingest a specification immediately' sentence preempts a likely conflation with other tools. Within the sibling set (breaking_changes, deprecation_scan, api_schema), this description unambiguously identifies the subscribe/watch responsibility, clearly differentiating it from one-off analysis tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides an explicit either/or usage decision: 'Without webhookUrl, read future events from the polling endpoint; with a public HTTP(S) URL, FactReason sends signed POST callbacks.' This gives strong contextual guidance for the tool's two modes. However, it does not literally name sibling alternatives (e.g., no 'use factreason_api_schema instead for...' phrasing), and the 'does not ingest' exclusion is slightly delayed rather than front-loaded as a warning.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.